A collection of CT images, manually segmented lungs and measurements in 2/3D

Tagsearth and nature, biology, image data, health conditions, healthcare

Context

Competitions like LUNA (http://luna16.grand-challenge.org) and the Kaggle Data Science Bowl 2017 (https://www.kaggle.com/c/data-science-bowl-2017) involve processing and trying to find lesions in CT images of the lungs. In order to find disease in these images well, it is important to first find the lungs well. This dataset is a collection of 2D and 3D images with manually segmented lungs.

Challenge

Come up with an algorithm for accurately segmenting lungs and measuring important clinical parameters (lung volume, PD, etc)

Percentile Density (PD)

The PD is the density (in Hounsfield units) the given percentile of pixels fall below in the image. The table includes 5 and 95% for reference. For smokers this value is often high indicating the build up of other things in the lungs.